Context Graph
Global Corpus Graph
Academe maps the global scholarly corpus into a connected graph so relevant work can be found through citation, method, author, topic, and evidence relationships.
What the graph contains
For each work, Academe tracks not just the text but the connections that make it meaningful. Every major scholarly discipline is covered, from AI and molecular biology to economic history and literary theory.
Papers, preprints, books, theses, datasets, patents, trials
Scholarly output sources Academe can legally reach, cleaned up so duplicate records do not crowd the interface.
Authors & research groups
Disambiguated across spellings, affiliations, and career moves.
Venues
Journals, conferences, and archives, with peer-review status and quality signals attached.
Concepts & topics
A layered vocabulary of tens of thousands of topics, so “transformer” in ML doesn’t collide with “transformer” in EE.
Methods, datasets & instruments
Papers using the same technique or dataset are connected even when they cite different ancestors.
Funders, institutions & grants
For when the question is “who is working on this, where, and on whose money?”
The connections that matter
What makes the corpus a graph and not a list is the edges between things:
- Citation edgesWho cites whom, with direction and context preserved: support, contradiction, extension, or critique.
- Co-authorship edgesWho works with whom, across institutions and across time.
- Topical edgesPapers that share methods, datasets, or underlying concepts, even when they live in different fields.
- Temporal edgesWhich ideas preceded which, so you can trace how a line of thinking evolved.
- Contradiction edgesExplicit disagreements, including replication failures, retractions, and methodological critiques, flagged rather than buried.
Why this matters for you
A list of search results is flat. A graph is navigable. When Academe surfaces a paper, it can tell you:
How it connects to your citations
Two hops through a shared method, a common co-author, or a cited dataset.
What it says about your argument
Whether it extends, supports, or contradicts a claim you’re already relying on.
What to read alongside it
The nearest nodes you’d need to understand its contribution.
What’s happened since
Challenges, replications, retractions, and adoption since publication.
Who else is in the neighbourhood
Active researchers working on the same question, with their recent output.
When it mattered most
Whether this is a recent result, a foundational reference, or somewhere in between.
How the graph stays current
The corpus is refreshed on a schedule. New papers, preprints, datasets, retractions, and corrections flow in so project context can reflect important changes in the literature.
Next